Which audience would be appropriate for calculating our response rate? And what is an adequate sample size?

@ChelseaIC there is no perfect science for this one, but as you rightfully call out the experience from those who purchased vs those who attended (with someone else purchasing) may have different experience.  So you should be cautious to conclude the experience is the same.  You would need to gather feedback from attendees and compare with ticket purchaser to see if the experience varies.

The right sample size, is typically a question of if the responses are statistically significant in representing the base.  The % who need to respond varies depending on the overall size of the base they represent.  Margin of Error is what you use to calculate this & to understand what is within your tolerance.  Refer to Your Guide to Margin of Error (With Calculator) - Qualtrics

5 & 15% response rate also isn’t a bad response rate, but you will see it’s a common question raised, very subjective, and no easy way to benchmark.  Our programs vary between 3 to 30% response rates as an example.  

If the volume of event attendees is quite small to begin with, you may also find you may not be able to get a statistically significant volume, as this is quite hard when you start with a small base.